Dysregulation of the peripheral and adipose tissue endocannabinoid system in human abdominal obesity

Abstract

The endocannabinoid system has been suspected to contribute to the association of visceral fat accumulation with metabolic diseases. We determined whether circulating endocannabinoids are related to visceral adipose tissue mass in lean, subcutaneous obese, and visceral obese subjects (10 men and 10 women in each group). We further measured expression of the cannabinoid type 1 (CB(1)) receptor and fatty acid amide hydrolase (FAAH) genes in paired samples of subcutaneous and visceral adipose tissue in all 60 subjects. Circulating 2-arachidonoyl glycerol (2-AG) was significantly correlated with body fat (r = 0.45, P = 0.03), visceral fat mass (r = 0.44, P = 0.003), and fasting plasma insulin concentrations (r = 0.41, P = 0.001) but negatively correlated to glucose infusion rate during clamp (r = 0.39, P = 0.009). In visceral adipose tissue, CB(1) mRNA expression was negatively correlated with visceral fat mass (r = 0.32, P = 0.01), fasting insulin (r = 0.48, P < 0.001), and circulating 2-AG (r = 0.5, P < 0.001), whereas FAAH gene expression was negatively correlated with visceral fat mass (r = 0.39, P = 0.01) and circulating 2-AG (r = 0.77, P < 0.001). Our findings suggest that abdominal fat accumulation is a critical correlate of the dysregulation of the peripheral endocannabinoid system in human obesity. Thus, the endocannabinoid system may represent a primary target for the treatment of abdominal obesity and associated metabolic changes.

FIG. 1

Circulating 2-AG concentrations in lean and obese subjects. Circulating 2-AG concentrations were measured in one lean control group (BMI <25 kg/m², n = 20) and in two obese groups (BMI >30 kg/m²) with different fat distribution phenotypes. Fat distribution was determined by measurement of abdominal adipose tissue areas using computed tomography. Subjects were divided in visceral (vis) obese (n = 20) or subcutaneous (SC) obese (n = 20) groups. A: 2-AG is clearly increased in subjects with visceral obesity. No sex differences were found for plasma 2-AG concentrations in the three groups. B: In both men and women, circulating 2-AG concentrations correlate with abdominal visceral fat area as measured by computed tomography scans. Data are means ± SE. *P < 0.05, **P < 0.01 by ANOVA.

FIG. 2

Circulating anandamide concentrations in lean and obese subjects. Circulating anandamide concentrations were measured in one lean control group (BMI <25 kg/m², n = 20) and in two obese groups (BMI >30 kg/m²) with different fat distribution phenotypes. Fat distribution was determined by measurement of abdominal adipose tissue areas using computed tomography. Subjects were divided in visceral (vis) obese (n = 20) or subcutaneous (SC) obese groups (n = 20). A: Anandamide (AEA) plasma concentrations differed between men and women. Anandamide levels are not significantly changed by obesity in women (B) and men (C). Data are means ± SE. *P < 0.05, **P < 0.01 by ANOVA.

FIG. 3

CB₁ and FAAH mRNA expression in lean and obese subjects. Gene expression was determined by real-time RT-PCR in paired samples of subcutaneous (SC) and visceral (vis) adipose tissue in all 60 subjects. A: In all groups, CB₁ mRNA expression was higher in visceral compared with subcutaneous adipose tissue (P < 0.05). Subjects in the lean group had significantly higher CB₁ mRNA expression than obese subjects, with no effect of the fat distribution phenotypes. B: In all groups, visceral FAAH mRNA levels were significantly higher than in subcutaneous adipose tissue. FAAH expression was highest in lean subjects and lowest in subjects with visceral obesity. Data are means ± SE. Group comparisons by ANOVA and t test for paired samples. *P < 0.05 for visceral vs. subcutaneous expression; #P < 0.05 between groups.

FIG. 4

Correlation between visceral and subcutaneous adipose tissue gene expression of CB₁ receptor (A) and FAAH (B) genes. Linear regression analysis with log-transformed data (1 AU = 1ag target gene/100 ng total RNA). Paired samples from visceral and subcutaneous adipose tissue were obtained from 60 Caucasian men (n = 30) and women (n = 30) and analyzed by real-time RT-PCR.

FIG. 5

Correlation between 2-AG plasma concentrations and insulin sensitivity. We found a significant correlation between 2-AG plasma concentrations and glucose infusion rate during the steady state of an euglycemic-hyperinsulinemic clamp. The correlation remained significant after adjusting for total body fat content. Data were log transformed to achieve normal distribution. % BF, percent body fat.

FIG. 6

Correlation between visceral CB₁ mRNA levels and SREBP-1c gene expression in visceral adipose tissue. Linear regression analysis with log transformed. Total RNA from visceral adipose tissue was obtained from 60 Caucasian men (n = 30) and women (n = 30) and analyzed by real-time RT-PCR.